Self-Supervised Visual Tracking via Image Synthesis and Domain Adversarial Learning

With the widespread use of sensors in applications such as autonomous driving and intelligent security, stable and efficient target tracking from diverse sensor data has become increasingly important. Self-supervised visual tracking has attracted increasing attention due to its potential to eliminat...

Full description

Saved in:
Bibliographic Details
Main Authors: Gu Geng, Sida Zhou, Jianing Tang, Xinming Zhang, Qiao Liu, Di Yuan
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/15/4621
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849239882129473536
author Gu Geng
Sida Zhou
Jianing Tang
Xinming Zhang
Qiao Liu
Di Yuan
author_facet Gu Geng
Sida Zhou
Jianing Tang
Xinming Zhang
Qiao Liu
Di Yuan
author_sort Gu Geng
collection DOAJ
description With the widespread use of sensors in applications such as autonomous driving and intelligent security, stable and efficient target tracking from diverse sensor data has become increasingly important. Self-supervised visual tracking has attracted increasing attention due to its potential to eliminate reliance on costly manual annotations; however, existing methods often train on incomplete object representations, resulting in inaccurate localization during inference. In addition, current methods typically struggle when applied to deep networks. To address these limitations, we propose a novel self-supervised tracking framework based on image synthesis and domain adversarial learning. We first construct a large-scale database of real-world target objects, then synthesize training video pairs by randomly inserting these targets into background frames while applying geometric and appearance transformations to simulate realistic variations. To reduce domain shift introduced by synthetic content, we incorporate a domain classification branch after feature extraction and adopt domain adversarial training to encourage feature alignment between real and synthetic domains. Experimental results on five standard tracking benchmarks demonstrate that our method significantly enhances tracking accuracy compared to existing self-supervised approaches without introducing any additional labeling cost. The proposed framework not only ensures complete target coverage during training but also shows strong scalability to deeper network architectures, offering a practical and effective solution for real-world tracking applications.
format Article
id doaj-art-67a46acd0b2e4b28a0287a592b29bd68
institution Kabale University
issn 1424-8220
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-67a46acd0b2e4b28a0287a592b29bd682025-08-20T04:00:49ZengMDPI AGSensors1424-82202025-07-012515462110.3390/s25154621Self-Supervised Visual Tracking via Image Synthesis and Domain Adversarial LearningGu Geng0Sida Zhou1Jianing Tang2Xinming Zhang3Qiao Liu4Di Yuan5Guangzhou Institute of Technology, Xidian University, Guangzhou 510555, ChinaSchool of Electrical and Information Engineering, Yunnan Minzu University, Kunming 650504, ChinaSchool of Electrical and Information Engineering, Yunnan Minzu University, Kunming 650504, ChinaSchool of Science, Harbin Institute of Technology, Shenzhen 518055, ChinaNational Center for Applied Mathematics, Chongqing Normal University, Chongqing 401331, ChinaGuangzhou Institute of Technology, Xidian University, Guangzhou 510555, ChinaWith the widespread use of sensors in applications such as autonomous driving and intelligent security, stable and efficient target tracking from diverse sensor data has become increasingly important. Self-supervised visual tracking has attracted increasing attention due to its potential to eliminate reliance on costly manual annotations; however, existing methods often train on incomplete object representations, resulting in inaccurate localization during inference. In addition, current methods typically struggle when applied to deep networks. To address these limitations, we propose a novel self-supervised tracking framework based on image synthesis and domain adversarial learning. We first construct a large-scale database of real-world target objects, then synthesize training video pairs by randomly inserting these targets into background frames while applying geometric and appearance transformations to simulate realistic variations. To reduce domain shift introduced by synthetic content, we incorporate a domain classification branch after feature extraction and adopt domain adversarial training to encourage feature alignment between real and synthetic domains. Experimental results on five standard tracking benchmarks demonstrate that our method significantly enhances tracking accuracy compared to existing self-supervised approaches without introducing any additional labeling cost. The proposed framework not only ensures complete target coverage during training but also shows strong scalability to deeper network architectures, offering a practical and effective solution for real-world tracking applications.https://www.mdpi.com/1424-8220/25/15/4621object trackingself-supervisedimage synthesisdomain adversarial learning
spellingShingle Gu Geng
Sida Zhou
Jianing Tang
Xinming Zhang
Qiao Liu
Di Yuan
Self-Supervised Visual Tracking via Image Synthesis and Domain Adversarial Learning
Sensors
object tracking
self-supervised
image synthesis
domain adversarial learning
title Self-Supervised Visual Tracking via Image Synthesis and Domain Adversarial Learning
title_full Self-Supervised Visual Tracking via Image Synthesis and Domain Adversarial Learning
title_fullStr Self-Supervised Visual Tracking via Image Synthesis and Domain Adversarial Learning
title_full_unstemmed Self-Supervised Visual Tracking via Image Synthesis and Domain Adversarial Learning
title_short Self-Supervised Visual Tracking via Image Synthesis and Domain Adversarial Learning
title_sort self supervised visual tracking via image synthesis and domain adversarial learning
topic object tracking
self-supervised
image synthesis
domain adversarial learning
url https://www.mdpi.com/1424-8220/25/15/4621
work_keys_str_mv AT gugeng selfsupervisedvisualtrackingviaimagesynthesisanddomainadversariallearning
AT sidazhou selfsupervisedvisualtrackingviaimagesynthesisanddomainadversariallearning
AT jianingtang selfsupervisedvisualtrackingviaimagesynthesisanddomainadversariallearning
AT xinmingzhang selfsupervisedvisualtrackingviaimagesynthesisanddomainadversariallearning
AT qiaoliu selfsupervisedvisualtrackingviaimagesynthesisanddomainadversariallearning
AT diyuan selfsupervisedvisualtrackingviaimagesynthesisanddomainadversariallearning